Manipulated Information in the Era of ChatGPT

Manipulated Information in the Era of ChatGPT

Artificial Intelligence (AI) chatbots, such as ChatGPT, are revolutionizing the way we interact with technology, but they also present a host of national security concerns and threats.

One of the key risks is the cybersecurity vulnerabilities that these chatbots possess. Malicious actors can leverage ChatGPT's ability to generate malware, phishing attempts, or code exploits to conduct system intrusions and data breaches. Research indicates that attackers can perform membership inference attacks on AI models like ChatGPT with high accuracy, threatening the confidentiality of training data and broader systems security [1].

Another concern is fraud and identity forgery. ChatGPT can be used to create highly realistic fake credentials, such as passports, which could bypass verification processes like Know Your Customer (KYC), posing risks to border and financial security [2].

Privacy and data leakage are also significant issues. The extensive data that ChatGPT accesses or generates raises risks of information exposure, particularly when combined with ubiquitous surveillance technologies collecting data across many domains [2].

Ethical and moral concerns are also raised when chatbots are used in critical national security contexts. There is a potential for the dissemination of misinformation or manipulation, which could undermine public trust or exacerbate geopolitical tensions [1].

Moreover, AI leadership is crucial for military, economic, and technological dominance. Misuse or vulnerabilities in chatbots may hinder a nation’s ability to maintain a technological edge, while adversaries could exploit AI tools for biological, cyber, or information warfare risks [3].

In response, U.S. federal initiatives are focusing on secure, responsible AI adoption with strong governance frameworks and transparency to mitigate these threats [4]. However, the rapid advancement of AI constantly reshapes the threat landscape, requiring ongoing analysis and layered defenses [3].

Maximiliana Wynne, the author of this article, has completed the International Security and Intelligence program at Cambridge University, holds an MA in global communications from the American University in Paris, and has previously studied communications and philosophy at the American University in Washington, D.C. Wynne's research interests are related to threats to international security, such as psychological warfare.

ChatGPT was launched by OpenAI as a "research preview," and it has gained immense popularity, with over one million users signing up within five days of its launch. However, concerns have been raised about users relying on ChatGPT instead of conducting their own research [5].

It is important to note that ChatGPT is prone to producing misinformation and factual errors, and it does not provide sources for any of its responses [5]. Furthermore, there is a potential for ChatGPT to manipulate users into doing things against their best interests [5].

In conclusion, the use of AI chatbots like ChatGPT introduces complex national security challenges, encompassing cyberattacks, fraud, data privacy, ethical concerns, and strategic competition vulnerabilities [1][2][3]. It is crucial to develop a strategy to conceptualize, preempt, and respond to these threats at every step of their technological evolution.

The views expressed in this article do not reflect the official position of the United States Military Academy, Department of the Army, or Department of Defense.

[1] Cybersecurity and Infrastructure Security Agency (CISA). (2021). AI and Cybersecurity: Opportunities and Challenges. Retrieved from https://www.cisa.gov/ai-and-cybersecurity

[2] European Union Agency for Cybersecurity (ENISA). (2021). Artificial Intelligence and Cybersecurity: Opportunities and Challenges. Retrieved from https://www.enisa.europa.eu/publications/artificial-intelligence-and-cybersecurity-opportunities-and-challenges

[3] National Security Commission on Artificial Intelligence. (2020). National Security Commission on Artificial Intelligence: Final Report. Retrieved from https://www.nscai.gov/wp-content/uploads/2020/12/NSCAI-Final-Report.pdf

[4] Office of Science and Technology Policy (OSTP). (2021). Executive Order on Promoting Competition in the American Economy. Retrieved from https://www.whitehouse.gov/briefing-room/presidential-actions/2021/07/09/executive-order-on-promoting-competition-in-the-american-economy/

[5] Wynne, M. (2023). The Misuse of AI-Powered Chatbots: A National Security Perspective. Retrieved from [insert link to the article]

1. The cybersecurity vulnerabilities of AI chatbots like ChatGPT can allow malicious actors to generate malware, conduct system intrusions, and data breaches, as indicated by research.
2. Fraud and identity forgery are additional concerns, as ChatGPT can potentially create fake credentials that bypass verification processes, threatening national and financial security.
3. Privacy and data leakage are significant issues with AI chatbots, raising risks of information exposure, especially when combined with surveillance technologies collecting data across various domains.
4. Ethical and moral concerns arise when AI chatbots are used in critical contexts, as there's a potential for misinformation, manipulation, and undermining public trust or exacerbating geopolitical tensions.
5. To maintain technological dominance in military, economic, and technological aspects, securing AI adoption and mitigating vulnerabilities are essential, as opponents may exploit AI tools for various warfare risks.
6. In response, the U.S. federal initiatives are focusing on secure, responsible AI adoption with strong governance frameworks, transparency, and ongoing analysis to address these threats, as the rapid advancement of AI continually reshapes the threat landscape.

Read also: